-
Notifications
You must be signed in to change notification settings - Fork 18
additional arm configs #21
Comments
+ @agners who might have a recommended armv6 config? |
If #20 works out, I don't think it should be too bad. |
ast2500_g5_defconfig is an armv6 machine (I'm the maintainer for that platform, so I have an interest in seeing it tested). It has a corresponding qemu machine |
@shenki yeah, sounds good then. We'll need a buildroot image for it. Have you worked with buildroot before? (I should sit down an write a blog post about it, otherwise I'll soon forget how to do so myself). |
On 11/3/18, Nick Desaulniers ***@***.***> wrote:
@tom-gall might have some more board specific configs we want make sure
continue to work.
I'm not sure if we want to get too wide with configs with travis, but maybe
this is feasible?
multi_v5_defconfig should be small, and covers some of the older platforms.
vf610m4_defconfig or stm32_defconfig would be good examples for ARMv7-M,
I'd suggest the former, since Stefan is familiar with that platform though the
stm32 one is more common.
Adding a big-endian config would also be helpful, as we've had problems
with that in the past.
Arnd
|
@shenki I guess that is the machine you created issue ClangBuiltLinux/linux#90 for? I guess something with ARMv6 architecture might be interesting anyway. Also RPi1 is ARMv6. There is ARM11MPCore emulation in Qemu, e.g. @arndb yeah I do have hardware to boot vf610m4_defconfig, actually I can boot such a kernel using U-Boot from the Cortex-A core of that SoC. Would be a good candidate for Kernel CI ARMv7-M testing... |
@tom-gall might have some more board specific configs we want make sure continue to work.
I'm not sure if we want to get too wide with configs with travis, but maybe this is feasible?
The text was updated successfully, but these errors were encountered: